878 



SCIENCE. 



[N. S. Vol. XXII. No. 574. 



The letter C in this formula stands for any- 

 color (in contrast to white, A = albinism) ; 

 G indicates gray, and Y yellow-bearing gam- 

 etes. The formula shows that 25 per cent, 

 should be pure yellow, 50 per cent, yellow- 

 grays (i. e., dominant recessives), and 25 pure 

 grays. To his surprise he got no pure yellow 

 mice, CYj but only impure yellows {%. e., 

 yellow dominants, GYCG-). Quantitatively 

 there were 72.7 per cent, yellows, CYCG, and 

 27.3 per cent. ' pure ' grays. 



Why were no pure yellows obtained? In 

 other combinations yellow seemed to follow 

 the same rule as the other colors, gray, black 

 or brown, but neither the pure yellow CY, 

 nor the albino with yellow alone recessive, 

 AY, nor the yellow heterozygote CYAY could 

 be obtained. 



Cuenot explains these curious results by 

 means of the assumption that the gametes 

 (eggs or sperm) having the formula CY or 

 AY never unite to give the forms CYGY, 

 AY AY, CYAY. In other words, yellow- 

 bearing spermatozoa never unite with eggs 

 bearing the yellow color alone. If this con- 

 clusion could be established with certainty it 

 would have a fundamental bearing on all 

 questions connected with Mendelian phenom- 

 ena. It is, therefore, of the utmost impor- 

 tance to know whether Cuenot's interpretation 

 is the only one that can be given to his results. 



It seems to me, in the first place, highly im- 

 probable that such a trivial difference as the 

 color of the hair could absolutely prevent the 

 conjugation of the gametes carrying these 

 colors, especially when we recall that the same 

 gametes must carry thousands of other unit- 

 characters that are identical. 



It seems also more probable that the re- 

 sults mean, that the color, yellow, was never 

 disassociated from one of the other colors, 

 gray, or black, or brown; and that the re- 

 sults are due to this and not to the disjunc- 

 tion of yellow and its failure to combine with 

 yellow again. On this basis, moreover, all the 

 facts can be explained according to the Men- 

 delian formula. Incidentally the whole ques- 

 tion of what is meant by the ' purity ' of the 

 germ cells is here raised. Let me now try to 

 show how these claims can be justified. 



Cuenot obtained his yellow mice by crossing 

 an albino of unknown ancestry with colored 

 mice. His yellow miice must, therefore, have 

 contained gray (or black). Now the theory 

 of ' pure ' germ-cells assumes, in order to ex- 

 plain the Mendelian ratio, that the germ-cells 

 of CYCG separate into two sorts, CY and GG^ 

 I assume, on the contrary, that the mixed 

 characters do not separate again, but alter- 

 nately dominate and remain latent giving 

 CY(GG) and (CY)CG. Two mice of this 

 kind paired will give: . 



CY(CG) (GY)CG 



CY{GG) (GY)GG 



IGY(GG) +2GY{CG) {GY)GG + 1 {GY)CG. 



Yellow Yellow Gray 



Thus there will be three yellows to one 

 gray. There are, moreover, two classes of yel- 

 lows, but while the first group CY(CG) will 

 breed true, the other group CY(CG)(CY)GG 

 will split in each successive generation accord- 

 ing to the Mendelian formula. There is an 

 implication in this point of view of importance 

 for the conception of germ-purity. The gray 

 in the latent condition in CY(CG) is differ- 

 ent from the gray in the recessive condition 

 in CY(CG)(CY)GG, for the former remains 

 in the latent condition to CY in inbreeding, 

 while the latter, the free CG, becomes domi- 

 nant in half of the germ cells. Thus the 

 germ cells of GYiCG) are all GY(GG), but 

 half the germ cells of CY(CG)(CY)GG are 

 yellow, CY(CG), and half gray, (CY)CG. 

 The same rule applies to the ' pure ' grays 

 (GY)CG in which the latent (CY) always 

 remains latent in inbreeding. Crossing with 

 other races may, however, bring the yellow 

 from its latent position to dominance or to 

 recessiveness again. 



Cuenot's yellow mice were, on my view, 

 either CY(CG) or CYiGG)(CY)CG. The 

 former if inbred should give usually yellow 

 mice, but if crossed with mice of other colors, 

 or united with GYiCG) (CY)CG, would give 

 some grays. Cuenot's somewhat ambiguous 

 statement on the top of page cxxix may ac- 

 cord with this view, although it may have a 

 different meaning. He obtained no pure 

 yellows, CYGY, because, on my view, his yel- 

 low mice were contaminated at the start. 



